JP6978084B2 - Control device, disk array device and patrol diagnostic method - Google Patents

Description

The present invention relates to a control device, a disk array device, and a patrol diagnostic method.

In the disk array device, patrol diagnosis is performed to detect the failure of the mounted drive. In patrol diagnosis, read commands and verify commands are issued to each drive at regular intervals, and it is confirmed whether an error response is returned. Patrol diagnosis needs to be performed for addresses in the entire area of the drive, but there is a problem that it takes time to complete the check of the entire area in a large-capacity drive such as that adopted in recent years.

To solve such a problem, Patent Document 1 provides an address generation circuit that generates an address for performing patrol diagnosis for each drive, and usually patrols while skipping the address based on the address generated by the address generation circuit. A method of performing a diagnosis and performing a patrol diagnosis by increasing the address by 1 when an error occurs is disclosed. By using this method, patrol diagnosis can be performed in a short time and an error can be detected at an early stage.

Special Publication No. 60-42504

The technique described in Patent Document 1 has a problem that a dedicated address generation circuit is required. Further, in recent years, from the viewpoint of high speed of data access, SSD (Solid State Drive) is increasingly used as a drive constituting a disk array device. There is a demand for a method for efficiently performing patrol diagnosis for SSD in a short time.

Therefore, an object of the present invention is to provide a control device, a disk array device, and a patrol diagnostic method that solve the above-mentioned problems.

According to one aspect of the present invention, it has a function of performing patrol diagnosis for a storage device, and when the storage device to be diagnosed is an SSD , a command is issued to the diagnostic unit that performs patrol diagnosis in block units and the storage device. , A command to acquire the type, identification information, and capacity of the storage device indicating whether the storage device is SSD or HHD, and if the storage device is SSD, further acquire the block size of the block constituting the SSD. , The identification information of the storage device, the type, the block size information of the SSD when the type is SSD, the capacity, and the configuration information acquisition unit for acquiring the block size information. A drive information storage table that stores the address to be diagnosed in association with each other, a product identification code table that stores the SSD identification information in association with the block size, and the identification information and the identification information in the drive information storage table. The control unit includes a control unit for registering the block size and the capacity, and the control unit includes the identification information acquired by the configuration information acquisition unit, the block size and the capacity, a predetermined address to be diagnosed, and the like. Is registered in the drive information storage table, and when the block size cannot be acquired by the configuration information acquisition unit, the product identification code table is referred to based on the identification information acquired by the configuration information acquisition unit. The block size corresponding to the identification information is acquired, the acquired block size is registered in the drive information storage table together with the identification information and the capacity acquired by the configuration information acquisition unit, and the diagnostic unit diagnoses. When the target storage device is an SSD, the block size corresponding to the identification information of the storage device to be diagnosed is read from the drive information storage table, and the patrol diagnosis of the SSD is performed in block units based on the block size. It is a control device that performs.

Further, according to one aspect of the present invention, it is a disk array device including a plurality of storage devices and the above-mentioned control device.

Further, according to another aspect of the present invention, it is a patrol diagnosis method executed by a control device having a function of performing patrol diagnosis for a storage device, in which a command is issued to the storage device, and the storage device is SSD. The block is obtained by acquiring the type, identification information, and capacity of the storage device indicating whether the storage device is HHD or HHD, and if the type is SSD, further issues a command to acquire the block size of the block constituting the SSD. The step of acquiring the size information, the identification information of the storage device, the type, the block size information of the SSD when the type is SSD, the capacity, and the address to be diagnosed are associated with each other. The block size is determined by the step of registering the identification information, the block size and the capacity acquired in the acquisition step, and the predetermined address of the diagnosis target in the drive information storage table to be stored, and the acquisition step. If it cannot be acquired, it corresponds to the identification information by referring to the product identification code table that stores the identification information of the SSD and the block size in association with each other based on the identification information acquired in the acquisition step. A step of acquiring the block size and registering the acquired block size in the drive information storage table together with the identification information and the capacity acquired in the acquisition step, and a case where the storage device to be diagnosed is an SSD. , The block size corresponding to the identification information of the storage device to be diagnosed is read from the drive information storage table, and the patrol diagnosis of the SSD is performed in block units based on the block size. The method.

According to the present invention, it is possible to shorten the execution time of patrol diagnosis for SSD and detect a failure at an early stage.

It is a schematic diagram of the information processing system which concerns on one Embodiment of this invention. It is a figure which shows an example of the drive information storage table by one Embodiment of this invention. It is a figure which shows an example of the SSD product identification code table by one Embodiment of this invention. It is a flowchart which shows an example of the registration process of the drive configuration information by one Embodiment of this invention. It is a flowchart which shows an example of the patrol diagnosis processing by one Embodiment of this invention. It is the first figure explaining the access to SSD. It is the 2nd figure explaining the access to SSD. It is a figure which shows the minimum structure of the control device in one Embodiment of this invention.

Hereinafter, patrol diagnosis for each drive constituting RAID (Redundant Array of Inexpensive Disks) according to an embodiment of the present invention will be described with reference to the drawings.
FIG. 1 is a schematic diagram of an information processing system according to an embodiment of the present invention.
As shown in the figure, the information processing system 100 includes a host device 1a, 1b, ~ 1m, a disk array device 1, and a drive group 40. The host devices 1a to 1m are connected to the disk array device 1. The disk array device 1 is connected to the drive group 40. The number of host devices connected to the disk array device 1 may be one. The drive group 40 includes a plurality of drives # 0 to # n. The plurality of drives # 0 to # n are storage devices such as HDD (Hard Disk Drive) and SSD (Solid State Drive), for example. Drives # 0 to # n have a redundant configuration by RAID. The drives # 0 to # n may be mounted inside the disk array device 1.

The host devices 1a to 1m are computers equipped with a processor such as a CPU (central processing unit). In the host devices 1a to 1m, the OS (operating system) is operated by the CPU. The OS outputs an IO instruction targeting the logical volume managed by the disk array device 1 to the disk array device 1.

The disk array device 1 configures and manages RAID by drives # 0 to # n of the drive group 40. Further, the disk array device 1 writes to the drive corresponding to the logical volume and reads from the drive according to the IO instruction from the host device 1a or the like. The disk array device 1 includes a host control unit 10, a cache memory 20, and a drive control unit 30. The host control unit 10 analyzes and executes commands received from the host devices 1a to 1m. The cache memory 20 stores various information. For example, the cache memory 20 includes a configuration information storage area 21 for storing configuration information of the drives of the drive group 40, and a data cache area 24 for storing data transmitted / received between the host devices 1a to 1m. The drive information storage table 22 and the SSD product identification code table 23 are stored in the configuration information storage area 21. These tables will be described later.

The drive control unit 30 controls the drive group 40. The drive control unit 30 includes a command issuing unit 31 and a diagnostic unit 33. The command issuing unit 31 issues a command to each drive included in the drive group 40. For example, the command issuing unit 31 issues a read command or a verify command to the address to be diagnosed at the time of patrol diagnosis. Further, the command issuing unit 31 includes an SSD information acquisition interface 32. The SSD information acquisition interface 32 issues a command to the SSD to acquire the configuration information of the NAND chip. The diagnosis unit 33 performs patrol diagnosis for each drive of the drive group 40. The patrol diagnosis is a process of inspecting whether or not a failure has occurred in the HDD or SDD. The diagnosis unit 33 determines an address to be inspected, and issues a read command or the like to the address via the command issuing unit 31. The diagnostic unit 33 confirms the response to the read command, and determines, for example, that the address is normal if there is a response to the command, and that the address is out of order if there is no response. The diagnosis unit 33 has a function of performing patrol diagnosis in block units when the drive is SSD.

Here, the patrol diagnosis for SSD will be described with reference to FIGS. 6 and 7.
6 and 7 are a first diagram and a second diagram illustrating access to the SSD, respectively.
The SSD is a storage device including a plurality of storage elements composed of a flash memory called a NAND chip. As shown, the storage area is composed of a plurality of blocks. The capacity of one block is 512KB or 1MB. Further, each block is composed of a plurality of cells. In SSD, once written data is written not in cell units but in block units. For example, in the example of FIG. 6, addresses of LBA (Logical Block Addressing) 1000 to 1400 are associated with block 0. Then, when the host device 1a or the like updates the data written in any of the addresses of LBA1000 to 1400, the data of the entire block 0 is rewritten instead of the cell in which the data is stored.

In addition, when one cell deteriorates and cannot be read or written, normally, the error correction code that the SSD controller has in the chip and the data of another cell are used to generate the data that cannot be read. , Copy the data to another block in block units and recover. However, if a plurality of cells deteriorate in the block, it cannot be recovered even with an error correction code. Then, it becomes impossible to read / write in block units, and it becomes impossible to read / write the data stored in the undegraded cell in the same block. Therefore, if it is checked whether the data can be read for any address in the block, it means that the check is also performed for another address in the block. That is, when the target drive for patrol diagnosis is SSD, an unrecoverable failure due to deterioration of a plurality of cells can be detected by performing a check on an arbitrary address in the block. In the present embodiment, the diagnostic unit 33 performs patrol diagnosis on the SSD in block units by utilizing this property. As a result, the execution time of the patrol diagnosis can be shortened as compared with the process of performing the diagnosis for all the addresses one by one. For example, in the case of the configuration example of FIG. 6, if diagnosis is made once for each of blocks 0 to block n (issue a verify command or the like and confirm the response), the diagnosis unit 33 has a range of LBA 1000 to 1400. It is possible to diagnose whether the address of is normal or abnormal.
The diagnosis in block units means that a diagnosis targeting one address is performed for each block, and the diagnosis is repeated for all blocks.

By the way, the size of the SSD block depends on the capacity and number of NAND chips, and therefore differs depending on the SSD product and generation. In order to perform diagnosis in block units, information on the block size of the SSD to be diagnosed is required. Therefore, for example, the command issuing unit 31 issues an INQUIRY command when the drive is connected, and acquires information such as the product identification code, type (HDD or SSD), and capacity of the drive. When the type is SSD, the command issuing unit 31 acquires block size information of the NAND chip from the SSD via the SSD information acquisition interface 32, and drives information stored in the configuration information storage area 21 of the cache memory 20. Record in the storage table 22. The diagnosis unit 33 executes block-based patrol diagnosis for SSD using the block size information recorded in the drive information storage table 22.

If the block size information cannot be acquired via the SSD information acquisition interface 32, the diagnostic unit 33 refers to the SSD product identification code table 23 stored in the configuration information storage area 21 of the cache memory 20 and commands the INQUIRY command. The block size information corresponding to the product identification code of the SSD drive obtained in the above is acquired. The diagnosis unit 33 executes a block-based patrol diagnosis for the SSD using the block size information read from the SSD product identification code table 23.

Next, the drive information storage table 22 and the SSD product identification code table 23 will be described.
FIG. 2 is a diagram showing an example of a drive information storage table according to an embodiment of the present invention.
As shown in FIG. 2, the drive information storage table 22 has fields for storing information of "product identification code", "type", "capacity", "patrol diagnostic address", and "NAND block size" for each drive. It is provided. For example, in the example of FIG. 2, the product identification code of the drive # 1 is "HDD80008", the type is "HDD", the capacity is "8TB", the patrol diagnostic address is "40000 sectors", and the NAND block size is "invalid". .. Similarly, the product identification code of drive # 2 is "SSD16000A", the type is "SSD", the capacity is "16TB", the patrol diagnostic address is "10000 sectors", and the NAND block size is "512KB".

The command issuing unit 31 issues an INQUIRY command to drive # 1 when the drive # 1 is connected, and issues a product identification code (“HDD80008”), drive type information (“HDD”), and drive capacity information (“HDD”). 8TB ”) is acquired, and the acquired information is associated with the drive # 1 and registered in the drive information storage table 22. When the drive type is "HDD", "invalid" is registered in the "NAND block size" field. Also, "0" is registered in the "Patrol diagnostic address" field. Then, when the patrol diagnosis is executed, the diagnosis unit 33 registers the value of the address to be diagnosed next in the "patrol diagnosis address" field. Similarly, when the connected drive is an SSD, the command issuing unit 31 stores the information acquired by issuing the INQUIRY command in the drive information storage table 22 in association with the drive. In the case of SSD, the size of the block, which is a unit for writing to the SSD, is registered in the "NAND block size" field as illustrated in drives # 2 and # 3 of FIG. At this time, if the connected SSD is an SSD corresponding to the SSD information acquisition interface 32, the command issuing unit 31 can acquire the block size by issuing the INQUIRY command. However, if the connected SSD is not the SSD corresponding to the SSD information acquisition interface 32, it may not be possible to acquire the block size by issuing the INQUIRY command. In such a case, the drive control unit 30 refers to the SSD product identification code table 23 described below based on the product identification code (such as SSD Product Identification information) acquired by issuing the INQUIRY command, and refers to the product identification code. Get the block size information corresponding to. Then, the drive control unit 30 registers the block size information acquired from the SSD product identification code table 23 in the "NAND block size" field of the drive information storage table 22.

Then, the diagnosis unit 33 executes the patrol diagnosis by using the information stored in the drive information storage table 22 when the drive is connected. For example, when the drive is an HDD, the diagnostic unit 33 refers to the drive information storage table 22 to acquire the value of the address from the "patrol diagnostic address" field of the drive to be diagnosed, and diagnoses the address. .. After the diagnosis, the diagnosis unit 33 adds the diagnosed sector size to the value in the "patrol diagnosis address" field. The value after addition indicates the next diagnostic address.

Similarly, when the drive is an SSD, the diagnostic unit 33 refers to the drive information storage table 22 to acquire the value of the address from the "patrol diagnostic address" field of the drive to be diagnosed, and makes a diagnosis for the address. .. After the diagnosis, the diagnosis unit 33 adds the block size value to the value in the "patrol diagnosis address" field. In the case of drive # 2 in FIG. 2, 512 is added to the diagnostic address 10000 this time. The value after addition indicates one address for diagnosis in the next block to be diagnosed next time.

Next, the SSD product identification code table 23 will be described.
FIG. 3 is a diagram showing an example of an SSD product identification code table according to an embodiment of the present invention.
As illustrated in FIG. 3, the SSD product identification code table 23 stores the SSD product identification code and the NAND block size (SSD block size) in association with each other. This information is a collection and registration of identification information and block size information about various SSD products of various generations provided or provided on the market. The SSD product identification code table 23 is stored in advance in the configuration information storage area 21 of the cache memory 20. With reference to the SSD product identification code table 23, if the SSD identification information is known, the information on the block size of the SSD can be acquired. Therefore, when performing the patrol diagnosis of SSD, the diagnosis can be performed in block units.

Next, a process of registering the SSD block size and the like in the drive information storage table 22 will be described.
FIG. 4 is a flowchart showing an example of a drive configuration information registration process according to an embodiment of the present invention.
As a premise, it is assumed that the drive information storage table 22 and the SSD product identification code table 23 are stored in the cache memory 20.
First, a new drive is inserted into the drive group 40. The drive control unit 30 detects the insertion of this drive (step S101). The drive control unit 30 instructs the command issuing unit 31 to acquire the drive configuration information. Then, the command issuing unit 31 issues an INQUIRY command to the drive in which the insertion is detected (step S102). The drive in which the insertion is detected responds to the INQUIRY command with information regarding the configuration of its own drive. The drive control unit 30 associates the product identification code, the drive type, and the drive capacity information with the drive in which the insertion is detected (for example, the slot number indicating the insertion position of the drive) from the returned information. And register it in the drive information storage table 22. Further, the command issuing unit 31 registers "0" in the "patrol diagnostic address" field of the drive information storage table 22 (step S103).

Next, the drive control unit 30 determines whether or not the type of the returned information is SSD (step S104). When the type is not SSD (step S104; No), the process of step S112 described later is performed. When the type is SSD (step S104; Yes), the drive control unit 30 instructs the command issuing unit 31 to acquire the SSD block size and the like. Then, the command issuing unit 31 issues a predetermined general SSD information acquisition command to the drive in which the insertion is detected via the SSD information acquisition interface 32 (step S105). The drive control unit 30 determines whether the response to this command is an error response (step S106).

When an error is responded to the SSD information acquisition command (step S106; Yes), the drive control unit 30 reads the product identification code of the drive previously acquired by the INQUIRY command from the drive information storage table 22. The drive control unit 30 refers to the SSD product identification code table 23 (step S107), and determines whether or not data matching the read product identification code is registered (step S108).

When the matching product identification code is registered (step S108; No), the drive control unit 30 reads out the value in the "NAND block size" field of the record in which the product identification code matches (FIG. 3), and the value thereof. Is registered in the "NAND block size" field of the data (record) in which the target product identification code is stored in the drive information storage table 22 (step S111).

When the matching product identification code is not registered (step S108; Yes), the drive control unit 30 sets "NAND block size" in the "NAND block size" field of the record in which the target product identification code is stored in the drive information storage table 22. "Invalid" is registered (step S109).

Further, when an error is not responded to the SSD information acquisition command (step S106; No), the drive control unit 30 sets the block size information acquired by the SSD information acquisition command as a target in the drive information storage table 22. The product identification code is registered in the "NAND block size" field of the record (step S110).

By the above processing, the drive control unit 30 completes the registration of the configuration information regarding the drive in which the insertion is detected in the drive information storage table 22 (step S112). Since the drive configuration information is registered when the drive is inserted in this way, all the configuration information related to drives # 0 to # n included in the drive group 40 is registered in the drive information storage table 22. In particular, when the drive is an SSD, the block size information is stored in the drive information storage table 22 unless the block size cannot be acquired. The diagnosis unit 33 uses this block size information to efficiently perform SSD patrol diagnosis.

Next, the flow of the patrol diagnosis process by the diagnosis unit 33 will be described.
FIG. 5 is a flowchart showing an example of patrol diagnosis processing according to an embodiment of the present invention.
For example, the diagnosis unit 33 shall perform patrol diagnosis for drives # 0 to # n a predetermined number of times in a predetermined cycle. The flowchart of FIG. 5 shows one of the patrol diagnostic processes. For example, the diagnostic unit 33 may perform the process described below a predetermined number of times in a predetermined cycle.
When the timing for performing the patrol diagnosis arrives, the diagnosis unit 33 first refers to the drive information storage table 22 (step S201) and acquires the diagnosis address (step S202). Specifically, the diagnosis unit 33 reads out the value of the "patrol diagnosis address" field in the drive to be diagnosed in the drive information storage table 22.

Next, the diagnosis unit 33 performs a diagnosis on the diagnosis address (step S203). Specifically, the diagnostic unit 33 instructs the command issuing unit 31 to issue a diagnostic command to the diagnostic address. The command issuing unit 31 issues a read command or a verify command to the instructed address. The diagnostic unit 33 acquires a response to the issued command and determines whether or not the diagnostic address is out of order. For example, if the diagnosis unit 33 responds to the command within a predetermined time, it determines that there is no failure. Further, for example, if there is no response within a predetermined time or an error code is returned, the diagnosis unit 33 determines that the address is out of order.

After completing the diagnosis for the current diagnosis address, the diagnosis unit 33 determines the address to be diagnosed next time. First, the diagnostic unit 33 refers to the drive information storage table 22 and acquires block size information. Specifically, the diagnostic unit 33 refers to the value in the "NAND block size" field in the drive to be diagnosed in the drive information storage table 22. The diagnosis unit 33 determines whether or not the block size is registered (step S204). The case where the block size is registered is the case where the drive is SSD and the block size can be acquired. The case where the block size is not registered is the case where the drive is an HDD or the drive is an SSD and the block size cannot be acquired.

(If the block size is known by SSD)
When the block size is registered (step S204; Yes), the diagnostic unit 33 reads the block size (value in the “NAND block size” field) from the drive information storage table 22 and acquires it (step S206). The diagnostic unit 33 advances the diagnostic address by the block address (step S207). For example, the diagnostic unit 33 adds the acquired block size to the diagnostic address this time. The value after addition is the address included in the next block. The diagnostic unit 33 updates the value in the "patrol diagnostic address" field with the added value. The updated value is the address to be diagnosed next time.

(When the block size is unknown on the HDD or SSD)
When the block size is not registered (step S204; No), the diagnostic unit 33 advances the diagnostic address by the diagnostic sector (step S205). For example, the diagnosis unit 33 adds the addresses of the sectors diagnosed this time to the diagnosis address of this time. For example, if the diagnosis for one sector is performed, the address of the next sector advanced by one sector is calculated. The diagnostic unit 33 updates the value in the "Patrol Diagnostic Address" field with the calculated value. The updated value is the address to be diagnosed next time.

If the patrol diagnosis address is larger than the drive capacity (value in the "capacity" field) of the drive information storage table 22 in the processes of steps S205 and S207, the diagnosis unit 33 puts the patrol diagnosis address in the "patrol diagnosis address" field. "0" is stored and the next diagnosis is performed from the beginning.

By the above processing, the diagnosis unit 33 completes the current diagnosis (step S208).
As described above, when the diagnosis target is SSD and the block size is known, the diagnosis unit 33 can perform the diagnosis in block units, so that each diagnosis can be performed in a short time. Further, since the diagnosis process only needs to be performed once for the representative address of the block, the diagnosis covering the entire area of the SSD can be completed in a short time. Further, even when the diagnosis target is SSD and the block size is unknown or the diagnosis target is HDD, the diagnosis can be performed in the same manner as the general patrol diagnosis, so that the diagnosis can be performed regardless of the drive type included in the drive group 40. , Patrol diagnosis can be performed.

If a disk array device that has multiple drives and manages them in a RAID configuration contains a large capacity drive, it takes time to diagnose all addresses, and there is a risk that a failure will occur before the diagnostic process is completed. There is. On the other hand, according to the present embodiment, when the drive includes an SSD, an area for acquiring and storing the configuration information of the NAND chip of the SSD is provided. Then, when performing patrol diagnosis, a process of diagnosing only one address for each block is performed for each block. As a result, even if the SSD is a large-capacity drive, the patrol diagnosis for the entire area can be efficiently completed in a short time. Therefore, it takes time to diagnose all addresses, which tends to occur in the case of a large-capacity drive, the risk of failure occurring before the diagnosis process is completed is reduced, and early failure detection becomes possible.

FIG. 8 is a diagram showing a minimum configuration of a control device according to an embodiment of the present invention.
As shown in the figure, the control device 200 includes at least a diagnostic unit 201.
The diagnosis unit 201 has a function of performing a patrol diagnosis on the drive (storage device). When the target storage device is SSD, the diagnosis unit 201 performs patrol diagnosis in block units. For example, when the SSD is composed of blocks 0 to block n, the diagnostic unit 201 performs a diagnosis on an arbitrary address of block 0 and determines whether or not block 0 is out of order. Then, the diagnosis unit 201 next performs a diagnosis once for an arbitrary address of the block 1 and detects a failure. Hereinafter, similarly, the diagnosis unit 201 arbitrarily selects an address representing each block for the blocks 2 to n, and makes a diagnosis only once for each block. As a result, even in a situation where the SSD has a large capacity and patrol diagnosis must be performed for a large number of SSDs, the diagnosis is significantly performed for all addresses one by one for the entire storage area. Diagnosis time can be reduced.
In comparison with the configuration of FIG. 1, the control device 200 corresponds to the drive control unit 30, and the diagnosis unit 201 corresponds to the diagnosis unit 33.

In addition, it is possible to replace the components in the above-described embodiment with well-known components as appropriate without departing from the spirit of the present invention. Further, the technical scope of the present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the spirit of the present invention.
For example, the diagnosis for SSD may be performed at least once (one address) for each block, but the number of times is not limited to one. Any number of times can be used as long as the effect of shortening the execution time of the patrol diagnosis can be obtained. For example, each block may be diagnosed 2-3 times. Further, the address for diagnosing the block is not necessarily limited to the value obtained by adding the block size to the address for which the diagnosis is performed this time. For example, it may be a value obtained by subtracting 1 from the value obtained by adding the block size to the address diagnosed this time. Further, the registration of the drive configuration information in the drive information storage table 22 by issuing the INQUIRY command is not limited to the time when the drive is inserted. The command issuing unit 31 is an example of a configuration information acquisition unit. Drives # 0 to # n are examples of storage devices.

100 ... Information processing system 1a, 1b, ~ 1m ... Host device 1 ... Disk array device 10 ... Host control unit 20 ... Cache memory 21 ... Configuration information storage area 22 ... Drive information storage table 23 ... SSD product identification code table 24 ... Data cache area 30 ... Drive control unit 31 ... Command issuing unit 32 ... SSD information acquisition interface 33, 201 ... Diagnosis unit 40 ... Drive group 200 ... Control device

Claims (8)

When the storage device to be diagnosed is SSD, it has a function to perform patrol diagnosis for the storage device, and a diagnostic unit that performs patrol diagnosis in block units .
A command is issued to the storage device to acquire the storage device type, identification information, and capacity indicating whether the storage device is SSD or HHD. If the storage device is SSD, the SSD is further obtained. A configuration information acquisition unit that issues a command to acquire the block size of the blocks that make up the block and acquires the block size information.
A drive information storage table that stores the identification information of the storage device, the type, the block size information of the SSD when the type is SSD, the capacity, and the address to be diagnosed in association with each other.
A product identification code table that stores the SSD identification information in association with the block size, and
A control unit that registers the identification information, the block size, and the capacity in the drive information storage table.
Equipped with
The control unit
The identification information, the block size and the capacity acquired by the configuration information acquisition unit, and the predetermined address to be diagnosed are registered in the drive information storage table.
When the block size cannot be acquired by the configuration information acquisition unit, the product identification code table is referred to based on the identification information acquired by the configuration information acquisition unit, and the block size corresponding to the identification information is acquired. Then, the acquired block size is registered in the drive information storage table together with the identification information and the capacity acquired by the configuration information acquisition unit.
When the storage device to be diagnosed is an SSD, the diagnostic unit reads the block size corresponding to the identification information of the storage device to be diagnosed from the drive information storage table, and blocks units based on the block size. Perform the patrol diagnosis of the SSD in
Control device. When the storage device to be diagnosed is an SSD, the diagnostic unit makes a diagnosis for the block based on a response to a diagnostic command issued to at least one address in the block, and makes a diagnosis for the block at the time of the next diagnosis. , Make a diagnosis for the other block based on at least one address in the other block.
The control device according to claim 1. When the storage device to be diagnosed is an SSD, the control unit adds the value of the block size to the address diagnosed this time, determines the address to be diagnosed at the time of the next diagnosis, and uses the address as the drive information. Register at the address to be diagnosed in the storage table,
The control device according to claim 2. If the storage device to be diagnosed is an SSD and the control unit cannot obtain the block size by referring to the product identification code table, the sector portion diagnosed this time is added to the address diagnosed this time. Then, the address of the diagnosis target at the time of the next diagnosis is determined, and the address is registered in the diagnosis target address of the drive information storage table.
The control device according to any one of claims 1 to 3. When the address related to the next diagnosis to be registered in the diagnosis target address of the drive information storage table is larger than the capacity, the control unit registers 0 in the diagnosis target address of the drive information storage table.
The control device according to claim 3 or 4. The diagnostic unit refers to the drive information storage table based on the identification information of the storage device to be diagnosed.
When the storage device is an HDD, the patrol diagnosis is performed by advancing the address to be diagnosed one sector at a time.
The control device according to any one of claims 1 to 5. With multiple storage devices
The control device according to any one of claims 1 to 6, and the control device.
A disk array device equipped with. It is a patrol diagnosis method executed by a control device having a function of performing patrol diagnosis for a storage device.
A command is issued to the storage device to acquire the type, identification information, and capacity of the storage device indicating whether the storage device is an SSD or an HHD. If the storage device is an SSD, the SSD is further configured. The step of issuing a command to acquire the block size of the block and acquiring the block size information, and
In the drive information storage table that stores the identification information of the storage device, the type, the block size information of the SSD when the type is SSD, the capacity, and the address to be diagnosed in association with each other. A step of registering the identification information, the block size and the capacity acquired in the acquisition step, and a predetermined address to be diagnosed.
When the block size cannot be acquired by the acquisition step, a product identification code table for storing the SSD identification information and the block size in association with each other based on the identification information acquired in the acquisition step is stored. A step of acquiring the block size corresponding to the identification information and registering the acquired block size in the drive information storage table together with the identification information and the capacity acquired in the acquisition step.
When the storage device to be diagnosed is an SSD, the block size corresponding to the identification information of the storage device to be diagnosed is read from the drive information storage table, and the SSD patrol is performed in block units based on the block size. Steps to make a diagnosis and
Patrol diagnostic method having.

Images